Free Lesson
Trace an ML Trading Strategy from Data to Backtest
30 min
Nov 12, 2026 11:00 AM
By continuing, you agree to Maven's Terms and Privacy Policy.
What you'll learn
Connect ETF Data, Model Forecasts, and Portfolio Results
Trace the Python workflow from time-aligned data and features to predictions, positions, and a backtest.
Check Validation and Trading Costs Before Judging a Result
Inspect walk-forward predictions, turnover, and costs to distinguish model scores from portfolio performance.
Use an AI Agent While Retaining Research Checks
See an agent implement a defined task, then check its data handling, experiment choices, and reported evidence.
Why this topic matters
A trading backtest reflects choices about data timing, features, validation, portfolio rules, and costs. We will trace an ETF example through those choices in Python, inspect where a result can mislead, and see where an AI agent can help run the work. Use the walkthrough to plan your own research, then continue in the December workshop or study at your own pace in Foundations.
